A prospective randomized study of the prevention of candida peritonitis (CP) in continuous ambulatory peritoneal dialysis patients using oral nystatin given concomitantly with antibiotic therapy was carried out for 2 years. Patients were randomized into two groups. Nystatin tablets 500,000 units four times a day were given to group 1 but not group 2 patients whenever antibiotics were prescribed. There were 199 patients at risk (mean follow-up, 18.0 months) in group 1 and 198 patients at risk (mean follow-up, 16.6 months) in group 2. The peritonitis and antibiotic prescription rates were comparable between the two groups. There were four episodes of CP in four patients in group 1 and 12 episodes in 11 patients in group 2. The probability of CP-free survival at 2 years was higher in group 1 compared with group 2 (0.974 v 0.915; P < 0.05). However, only three (75%) CP episodes in group 1 and six (50%) in group 2 were considered "antibiotics related." The incidence of antibiotics-related CP was 1.39 and 3.19 per 100 peritonitis episodes and 0.66 and 1.43 per 100 antibiotic prescriptions in groups 1 and 2, respectively (P = NS). We conclude that oral nystatin prophylaxis with each antibiotic prescription reduced the rate of CP in patients on continuous ambulatory peritoneal dialysis irrespective of its apparent temporal relationship to antibiotic prescription.